Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title Exploring Statistical Inference from Web-Based Surveys: How Does It Work?
Author Santos, R.
Year 2007
Access date 25.05.2007
Abstract

Web-based surveys are increasing in popularity because of their cost effectiveness and their ability to generate survey results quickly. Two principal web-based designs have emerged in the first have of this decade: (1) panels recruited from household RDD samples that subsequently employ web technology for gathering survey data; and (2) opt-in panels of internet users whose survey data are adjusted via extant data (e.g., an independent RDD telephone survey). The purpose of this paper is to provide a framework for easily understanding the source of statistical theory used to generate inferences about the survey population. A simple expression of the bias of the mean is used to show how classical finite population sampling theory is no longer the primary driver of statistical inference and that adoption of model based theory is needed for both the web-based designs under investigation. But the models that underlie the validity of inference differ for the two web-based designs - one relies on the veracity of a nonresponse model, while the other relies on a model for noncoverage. The paper concludes with thoughts on how our industry might develop empirically based, defensible models in order to justify (or dispel} the validity of current web-based designs.

Access/Direct link

Conference homepage

Year of publication2007
Bibliographic typeConferences, workshops, tutorials, presentations
Full text availabilityAvailable on request
Print

Web survey bibliography - The American Association for Public Opinion Research (AAPOR) 62th Annual Conference, 2007 (48)